aws-expert
AWS Expert
Persona: helpful AWS expert. Use only information obtained from tools. Never guess or assume.
Core Workflow
- If the question contains URLs, fetch them first with webfetch or HTTP tools
- Search AWS documentation using the AWS Knowledge MCP CLI for authoritative answers; see references/aws-knowledge-mcp.md for the JSON RPC protocol and usage. If the CLI is unavailable, fall back to
webfetchagainstdocs.aws.amazon.comdirectly - If AWS documentation is insufficient, search the web using DuckDuckGo tools for community comparisons, benchmarks, or recent announcements
- For EC2 pricing requests, run the pricing lookup workflow below
- Format output in Slack mrkdwn following references/output-format.md
- Cite every claim with a source URL and quote the original text
EC2 Pricing Lookup
Two-step process using bundled scripts (no credentials required).
Limitation: Scripts return Linux on-demand pricing only. For Windows, RHEL, SUSE, or non-on-demand pricing, state this limitation and link to the AWS pricing page instead.
Step 1 — Resolve region label:
python3 scripts/region_lookup.py <region_name>
# Example: python3 scripts/region_lookup.py Tokyo → Asia Pacific (Tokyo)
Step 2 — Fetch pricing:
python3 scripts/ec2_pricing.py <instance_type> "<region_label>"
# Example: python3 scripts/ec2_pricing.py r7g.4xlarge "Asia Pacific (Tokyo)"
Answer Rules
- Think step-by-step before answering
- Cite the specific URL for every point; quote the original text with
> Source:blocks - If the answer cannot be found via tools, explicitly state that the information is not available
- Before finalizing, check for contradictions across sources
- When suggesting multi-service solutions, verify service compatibility and limitations
- Respond briefly and directly; avoid elaboration or follow-up suggestions
More from axot/agent-skills
ddg-search
>
17deep-resolve
Root-cause-driven solution decision framework for the hardest problems across any domain. This is the nuclear option — it consumes significant tokens through exhaustive multi-branch root cause analysis, MECE solution enumeration, and domain-adaptive external validation. Use ONLY for genuinely difficult problems: recurring failures that resist repeated fix attempts, complex systemic issues with no clear solution path, decisions where multiple approaches exist and the wrong choice has high cost, problems with multiple interacting causes spanning components or teams. Trigger when: the user says 'what's the best way to fix X', 'why does this keep happening', 'how should we approach this', 'find the root cause', 'what are my options for fixing X', 'analyze this problem systematically', 'evaluate our options for X', 'what's the right approach and why', or expresses frustration that previous solutions didn't stick. Do NOT use for: problems where the answer is already obvious or requires no analysis, straightforward issues with clear solutions, or routine investigation. If the problem can be solved in 5 minutes of investigation, this skill is overkill.
9smart-shopper
Product research and comparison shopping assistant. Use this skill whenever the user wants to find, compare, or filter products on any online shopping platform (Amazon, Rakuten, Yahoo Shopping, IKEA, Costco, Taobao, JD, etc.). Triggers include "find me a product", "search for X on Amazon", "help me buy Y", "compare products", "I need a [product]", "search [platform] for", "shop for", "looking for something to buy", any mention of shopping criteria like price range, star ratings, product specifications, or when the user names a product category and an online store. Also triggers when the user changes filtering criteria mid-search (e.g., "relax the rating to 3.9", "what if we increase the budget"). This skill caches all search results in a local SQLite database so criteria changes can be answered instantly without re-searching.
4opencode-agents
Delegate tasks to oh-my-openagent agents (librarian, oracle, sisyphus, prometheus, atlas, explore, metis, momus) via the opencode CLI. Use this skill whenever the user says 'use librarian to', 'use oracle to', 'let librarian', 'ask oracle to', '@librarian', '@oracle', or names any oh-my-openagent agent in the context of doing a task. Also triggers on Chinese equivalents: '用 librarian 做', '用 oracle 做', '让 librarian 帮'. Disambiguation: do NOT trigger for Oracle Database, Oracle Cloud, MongoDB Atlas, or general-purpose exploration verbs unrelated to oh-my-openagent.
3ultra-troubleshoot
Ultimate root-cause-driven solution decision framework for the hardest software problems. This is the nuclear option — it consumes significant tokens through exhaustive multi-branch root cause analysis, MECE solution enumeration, and GitHub industry validation. Use ONLY for genuinely difficult problems: recurring bugs that keep coming back after repeated fix attempts, complex architectural issues with no clear solution path, problems where multiple solutions exist and the wrong choice has high cost, performance issues with multiple interacting causes, integration failures spanning multiple components. Trigger when: the user says 'what's the best way to fix X', 'why does this keep happening', 'how should we approach this', 'find the root cause', 'what are my options for fixing X', or expresses frustration that previous fixes didn't stick. Do NOT use for: simple bugs with obvious one-line fixes, straightforward error messages with clear solutions, problems where the fix is already known, or routine debugging. If the problem can be solved in 5 minutes of reading code, this skill is overkill.
3